R Notebook

localData = read.csv('local1Data.csv')
localData = localData[,2:ncol(localData)]

head(localData)

colnames(localData)

 [1] "networkName"         "methodName"          "nodeList"            "andNodeList"        
 [5] "andNodeInvertList"   "ruleLengths"         "equivs"              "plainRules"         
 [9] "randomERSIndividual" "minLocalSearchError"

getIndegree = function(x){
  return(stringr::str_count(pattern = "\\[",string = x) - 1)
}
localData$indegree = unlist(lapply(as.vector(localData$equivs), getIndegree))

require(plyr)
levels(localData$methodName) = c("Combined", "Phosphoproteomics", "Proteomics", "Transcriptomics")
mu <- ddply(localData[localData$indegree>=3,], "methodName", summarise, grp.mean=mean(ruleLengths))
mu$grp.mean = round(mu$grp.mean,1)
ggplot(data = localData[localData$indegree>=3,]) + geom_histogram(aes(ruleLengths), bins = 20) + theme_bw() + theme(panel.grid.minor = element_line(color = 'grey', size = 0.25, linetype = 2), text = element_text(size = 14))  + ylab("Count") + scale_x_continuous(name = "Size of ERS") + facet_wrap(vars(methodName), nrow=1, scales="free_y") + geom_vline(data = mu, aes(xintercept=grp.mean), color='red',linetype='dashed') + geom_label(data = mu, y = Inf, x=Inf, vjust=2, hjust=2, aes(label=grp.mean), color = 'red') + facet_wrap(vars(methodName), nrow=1, scales="free_y")
ggsave(filename = "ERS_size_distribution.png", height = 4, width = 8, dpi = 600)

ggsave(filename = "ERS_size_distribution.svg", height = 4, width = 8)
ggsave(filename = "ERS_size_distribution.pdf", height = 4, width = 8)

methodName	grp.mean
concatenated	44.15706
Phosphoproteomics	43.71438
Proteomics	56.88941
Transcriptomics	62.70384

nodeTable = read.csv("node_table.csv")
nodeTable = nodeTable[,2:ncol(nodeTable)]
print(colnames(nodeTable))

 [1] "andNode"      "Display.Name" "IS"           "name"         "RA"           "selected"    
 [7] "shared.name"  "Dataset"      "Contrast"     "Network"